Clinical presentation, diagnostics, and outcomes of infants with congenital and postnatal tuberculosis: a multicentre cohort study of the Paediatric Tuberculosis Network European Trials Group (ptbnet).

According to estimates, globally more than 200,000 pregnant women develop tuberculosis (TB) annually. Despite this, data on perinatal TB remain scarce. This study aimed to describe perinatal TB, comprising congenital (cTB) and postnatal (pTB) TB, in a European setting. Retrospective cohort study via the Paediatric Tuberculosis Network European Trials Group (ptbnet) capturing and comparing cases of cTB and pTB diagnosed at 104 participating European healthcare institutions between 1995 and 2019. Forty-six cases reported by 20 centres were included in the final analysis (cTB, n = 27; pTB, n = 19). Median age at symptom onset was one week in cTB (IQR: 0-1 weeks), and 12 weeks in pTB patients (IQR: 5-18 weeks). Prematurity was more common in cTB than pTB patients [57.9% (11/19); 95% CI: 36.3-76.9% vs. 21.1% (4/19); 95% CI: 8.5-43.3%; p = 0.049], and the average birth weight was significantly lower [1680 g; IQR: 932-2805 g vs. 2890 g; IQR: 2461-3400 g; p = 0.0043]. Microbiological confirmation was achieved in most patients [85.2% (23/27); 95% CI: 67.5-94.1% vs. 78.9% (15/19); 95% CI: 56.7-91.5%; p = 0.70]. The sensitivity of interferon-gamma release assays was poor in both groups [25.0% (3/12) 95% CI: 8.9-53.2% vs. 35.7% (5/14) 95% CI: 16.3-61.2%; p = 0.68]; in contrast, the sensitivity of the tuberculin skin tests (at 5 mm cut-off) was significantly higher in pTB patients [16.7% (2/12) 95% CI: 4.7-44.8% vs. 66.7% (10/15); 95% CI: 41.7-84.8%; p = 0.0185]. Approximately half of the patients required intensive care support [51.9% (14/27) 95% CI: 34.0-69.3% vs. 47.4% (9/19); 95% CI: 27.3-68.3%; p > 0.99]. Four (4/46; 8.7%) patients died, and four (4/46; 8.7%) had severe long-term sequelae. There was substantial mortality and morbidity in this patient cohort, despite the high-resource setting. cTB was associated with premature birth and low birth weight. In contrast to microbiological tests, immunological tests perform poorly in perinatal TB, and should therefore not be used as rule-out tests. No study-specific funding.